System Performance

Pixel phones have been known to be among the best performing Android devices in the market. This is mainly due to the Pixel’s performance team taking the time and attention to tweak the software stack – kernel and userspace alike. This is one of the benefits of being one of the last flagships out of the gate for a given generation, as it gives time to optimize the performance. The Pixel 3 comes with the Snapdragon 845, and I’ve written many times this year how Qualcomm’s software, and in particular the kernel scheduler was a very significant factor as to why this year’s Snapdragon phones performed so marvellously.

One of the big questions I posed myself early in the year is exactly how Google planned to handle Qualcomm’s great divergence from upstream, and the divergence from the Google common kernel. As a reminder, the Google common kernel is now the “official” branch on which SoC vendors should be basing their BSP (board support packages, essentially the software stack) for their own products. This is a collaborative effort between vendors (Mainly Google, Qualcomm and Arm), and it’s also the target where Arm pushes its own EAS patches.

The matter of fact is, for the Pixel 3, Google is simply using Qualcomm’s custom scheduler. This is both a great win for Qualcomm given the expected device performance of the Pixel 3, and quite a blow to Arm’s own efforts, as the EAS improvements over the last year are just simply not being used. Qualcomm’s efforts as well as the resulting product are just too good to pass on, and I’m very much expecting next year to finally be a watershed moment where other vendors finally abandon attempts to keep things minimalistic, and in line with upstream Linux, and finally see the immense value in investing in actual immediate benefits for consumer devices of a given generation.

PCMark Work 2.0 - Web Browsing 2.0

Starting with PCMark’s Web Browsing 2.0 test, the Pixel 3 leads the pack, with a slight advantage over other Snapdragon 845 phones. The reason here is that Google seemingly uses the most up-to-date scheduler, as well as has some possible file I/O advantages which I’ll get into a bit later. There are also possible OS side improvements in the libraries, as the Pixel 3’s ship with Android 9.

I’ve updated the performance results for past Pixels with the newest OS updates, as well as for devices like the OnePlus 6 as these have received their OS updates as well.

PCMark Work 2.0 - Video Editing PCMark Work 2.0 - Writing 2.0

In the writing test, which is probably PCMark’s most important as well as representative benchmark, the Pixel 3 saw a big leap in performance over the previous Pixels – however I think this was due to Android 9 itself, as we also saw a big jump in the OnePlus 6’s performance with the latest OS update.

PCMark Work 2.0 - Photo Editing 2.0

The photo editing test is very much a scheduler responsivity test as modern devices are able to complete the workloads relatively fast at their peak performance states. Here the score wildly fluctuates depending on how fast the DVFS mechanism is, and we see the Pixel 3 among the best performers.

PCMark Work 2.0 - Data Manipulation

The data manipulation score is extremely high on the Pixel 3 compares to other phones, including the OnePlus 6. I wasn’t able to verify this empirically, but glancing over the scheduler the Pixel has some unique updates to it which facilitate better responsiveness and scheduling of single big tasks, and the data manipulation test is such a workload with a big single-threaded component.

PCMark Work 2.0 - Performance

Overall, the Pixel 3 takes the top position in PCMark, all thanks to its scheduler improvements as well as a slight advantage due to it running Android 9.

Speedometer 2.0 - OS WebView

Moving onto web browser tests, the Pixel 3 largely matches the other Snapdragon 845 devices. This is no surprise as Speedometer 2.0 is a high constant throughput ST benchmark, and as such isn’t as affected by scheduler as PCMark.

Apple still has a considerable performance lead here. After our recent iPhone XS review and SoC deep-dive, I’m more leaning towards the explanation that a big part of the advantage here is purely due to hardware and the microarchitectural advantages of Apple’s CPUs, with part of it also being Apple’s Nitro JS engine.

WebXPRT 3 - OS WebView

WebXPRT also looks in line with other Snapdragon 845 devices.

Pixel 3 – Now using F2FS

Section with credit and input by Park Ju Hyung (@arter97)

The Pixel 3 now has switched over from an EXT4 filesystem, to the F2FS filesystem. Google explains this switch due to the fact that F2FS now supports inline block encryption which has been the last major roadblock as to why Google hadn’t made the switch earlier.

Inline block encryption uses the SoC’s inline cryptographic engines, which just serve as an intermediate hardware layer to the NAND and offload any encryption workloads that were initially in past devices performed by the CPU.

The switch to F2FS now gives the Pixel 3 a number of advantages over previous filesystem; Previously, SQLite (which is used by almost all database files under Android) used another 'journaling' on its own to prevent corruption. This caused “double journaling” on top of EXT4, which in itself is a journaling filesystem. Since F2FS doesn’t need this kind of protection and the Pixel 3 includes Google’s SQLite changes in Android 8.1, the Pixel 3 is able to take advantage of this, as well as any other F2FS based device from other vendors which have the corresponding OS patches.

The result is that this will enable much higher write/commit speeds for SQLite, not to mention less wear and tear to the underlying UFS storage. Also, the Pixel 3 turned off barriers for fsync() system calls, which will improve general random I/O write speeds by a significant margin.

Another big improvement for file I/O is the implementation of “Host Performance Booster” in the kernel and UFS controller firmware stack. HPB is essentially caching of the NAND chip’s FTL (flash translation layer) L2P (logical to physical) mapping tables into the hosts (SoCs) main memory. This allows the host driver to look up the target L2P entry directly without betting on UFS’s limited SRAM to have a cache-hit, reducing latency and greatly increasing random read performance. The authors of the feature showcase an improvement of 59-67% in random I/O read performance due to the new feature. It’s worth to mention that traditional Android I/O benchmarks won’t be able to show this as as those tend to test read speeds with the files they’ve just created.

Overall, the Pixel 3 is the fastest Android device on the market right now. The one thing that puts it above other devices such as the OnePlus 6 is a noticeable faster response-time when opening applications – either a framework related boost or just an effect of the faster file I/O.

Introduction & Design GPU Performance
POST A COMMENT

134 Comments

View All Comments

  • Impulses - Friday, November 02, 2018 - link

    Meh, their memory management is still over-aggressive here... Some of the instances I've seen of apps getting killed after opening the camera don't even happen on my OG Pixel with the same amount of RAM. Reply
  • Arbie - Friday, November 02, 2018 - link

    Is anyone actually in favor of not having a headphone jack?

    I get it that you *can* make the phone very slightly thinner without one; OTOH keeping the same thickness *can* permit longer battery life, which everyone wants. I also understand that you *can* put a superlative DAC in a dongle, for audiophiles. But that can be done anyway.

    Killing the jack seems to bring no benefit to the majority of folks, and is going to alienate a lot of them. So - why? Such a change can't just be due to fashion.
    Reply
  • zanon - Friday, November 02, 2018 - link

    Yes, and it blows my mind that so many so-called tech enthusiasts are actually actively against improvements to archaic technology. Wires suck, they get snagged on things, they limit movement, they act as antennas that pick up things like GSM sync signals, etc. We used them because that was the best that could be done in the 70s and 80s. But that's not the case now. They offer literally zero inherent advantages except one: the antenna part, they can in principle (and in a few phones over the years) act as physical antennas for classic radio broadcasts. But that simply hasn't been that motivating to the developed world market in a long time, as evidenced by the fact that it's not a major advertised universal Android feature and Apple never bothered with it at all.

    Beyond that wireless should be fine, and I think people confuse implementation trouble due to companies doing a bad job or market immaturity with real problems. Audio quality is not an issue, Bluetooth 5 has plenty of bandwidth for even lossless audio (and 256 AAC let alone Sony's fat codec is transparent anyway). Run time is fine, once something gets to 10-18 hours+ on a charge that meets what I'd ever listen to in a single session with no breaks (when it could be charged). Symmetric encryption is cheap and easy now and means it's as secure (or more given TEMPEST) as wired. Some stuff is still lacking, like easy switching between multiple wireless sources, but that's not a fault of the inherent tech so much as there hasn't been any real demand for it yet because wires have hung on so damn long. We're already seeing way more cool adapters for existing products get released due to increased demand and so I expect the market to work it's typical magic.

    I'm delighted to see that relic go. I have no nostalgia towards it anymore then I do towards 8-track or MMX or 10base ethernet or whatever.
    Reply
  • cfenton - Friday, November 02, 2018 - link

    "Beyond that wireless should be fine, and I think people confuse implementation trouble due to companies doing a bad job or market immaturity with real problems."

    But those are real problems for everyday use. Just because the technology is theoretically there doesn't mean it works well in practice. I use wireless headphones with my Galaxy S7 in the winter because I hate threading a cord through my coat. About 20% of the time my headphones only connect for voice calls, but not media. I have to either fiddle with menus, or turn them off and on again. I'm using a $100+ over-ear set, not some cheap $20 ear-buds from a no-name brand. It's annoying enough that I still use my wired headphones at home, and outside in the summer.

    I've heard Apple's solution is a lot better, but I don't have an iPhone, and don't particularly like Beats.

    If getting rid of the headphone jack added anything to the phone, or somehow made bluetooth better, I'd be all for it. But it seems to add nothing. Samsung has shown with the Galaxy S9 that you can make a very nice modern phone and still have room for a headphone jack. So why not have both until the real world problems with wireless get worked out?
    Reply
  • zanon - Friday, November 02, 2018 - link

    But how do you think we have *ever* gotten from A to B there? I'm only in my late-30s so not really old, but even so I can still remember most of the PC era and I can't remember any significant technology that didn't take many years to get refined, to fill out niches, to unambiguously beat what it replaced. Not just in hardware but often even in software, take audio and video codecs for example, there have been a number of replacement cycles where the technically superior standard spec for a while was inferior in practice to highly refined encoders for the previous generation.

    At some point somebody has to get the ball rolling and there will be a few generations where people just have to deal with compromises. The only way we've ever gotten great tech is for a real market to get established and then iterated upon. In the case of BT audio, in just the last year I've seen more improvements in things like BT adapters then like the previous 5 years at least. Just a few days ago a bunch of BT5 gen 2 refined ones came out at lower prices and better performance.

    >"So why not have both until the real world problems with wireless get worked out?"

    Because that has never, ever worked. If the old thing is still available major players just use the old thing. Inertia is very powerful. Somebody sometime has to bite the bullet, and given the realities of mass manufacturing you can't "have all the problems worked out" in version 1.0. Version 1.0 is always going to have issues. But you can't get ver2 or ver3 without going through 1.0 first. In this case Apple has shown it can be done very well and with more vendors going that way there is now a clear market of people ready to spend money on it and 3rd parties are starting to react and iterate. Now we can look forward to having way better stuff in another year or 2, whereas if they had all put it off then there is no reason to expect that'd be the case anymore then it was the last decade.

    If you really don't want to be on the bleeding edge there then no problem, just get an older device! Mobile lasts a lot longer now, Apple already supports their stuff 5+ years and Google is getting better too. But those of us who like to live nearer the edge have always had to deal with that edge being rougher then those who follow. Somebody has to go first though.
    Reply
  • porcupineLTD - Saturday, November 03, 2018 - link

    Nice mental gymnastics to justify an idiotic trend. Reply
  • zanon - Saturday, November 03, 2018 - link

    Nice brainless technophobia from a luddite lol. Why don't you go back to your retro forum hole and cling to your VGA and CRT claiming they're the best ever and all this flatscreen stuff is a sheeple fad? Reply
  • Impulses - Saturday, November 03, 2018 - link

    CRT still haven't been surpassed for some fast refresh applications... And likely never will by LCD, OLED might get there. Reply
  • cfenton - Saturday, November 03, 2018 - link

    But we're not on version 1.0 of Bluetooth, we're on version 5.0 and it's still not great. Well, I don't have any 5.0 stuff, so maybe they finally got it right, but version 4.0 is still pretty inconsistent. If wireless worked as well as wired, I'd be all for it, but in my experience, it doesn't yet.

    As for transitions not happening unless forced, I'm not sure that's true. Lots of people use Wi-Fi even though wired ethernet still exists. It's only in the last five years or so that laptops have been shipping without ethernet ports, and desktops still have them. Again, I'm not against having good wireless tech, I just don't think it's 100% there yet and removing wired connections doesn't seem to add anything to the phone.
    Reply
  • zanon - Saturday, November 03, 2018 - link

    >"But we're not on version 1.0 of Bluetooth, we're on version 5.0 and it's still not great."

    It is great (or at least getting close) from some vendors, which shows that the technical foundation is there. Once that exists the only real way to have progress happen sustainable is to have a market for it so hundreds to thousands of attempts will be made, most of which will be mediocre, but some of which will be good and then get copied/followed in turn. To your own point:

    >"Lots of people use Wi-Fi even though wired ethernet still exists."

    WiFi was introduced *20 years ago*. Yes, 1998. And it stunk. It was a long time before it even achieved the kind of technical spec it needed, 802.11n took 11 years, 802.11ac took 15 years. Both came after smartphones and notebooks overtaking desktops, demand drove innovation and pricing. Of course, even now in 2018 a lot of WiFi stuff on the market is still junk and will still give a bad experience, but at least there are good options.

    I think a better example would be USB replacing PS2 & ADB ports. Wow were there a lot of howls over that. We're seeing some of the exact same thing now in fact with USB-C and TB3, and people raging about adapters and "everything was fine before" and the early pains of these standards (plenty of junk USB-C implementations). But the fact is USB-C is a nice connector that solves physical and speed problems.

    >"I just don't think it's 100% there yet and removing wired connections doesn't seem to add anything to the phone."

    I absolutely agree it's not 100% there yet! There is clearly a ways to go. But I also don't think it'll ever go from 20% or even 50% to 100% in one leap either, no matter when they started the first few versions would have compromises. But I've seen more progress in the last year then the last decade, and once people know that something can be done well it tends to drag up the industry because customers are less willing to accept excuses of "oh nobody can do this."
    Reply

Log in

Don't have an account? Sign up now